An individual's genetic background plays an important role in their susceptibility to cancer, disease progression, and response to therapy. Genetic variations can be used as markers to understand complex diseases including cancer, and the immune/inflammatory response. The advent of the sequence of the human and other genomes, the availability of microarrays, and our ability to type genetic markers and sequence whole genomes has led to the need for enhanced computing and informatics. Inflammation is recognized as playing an important role in cancer. However the extent and implications of genetic variation in human genes involved in inflammation is poorly characterized. We are interrogating the germline gene variation in cancer patients, the genetic variation of tumor cells and the variation of genes involved in the inflammatory response. Prostate Cancer Analysis of the MSMB protein A series genome-wide association studies (GWAS) performed at the NCI revealed a set of loci genetically associated with prostate cancer. One of these loci include the gene for MSMB, a gene that encodes microseminoprotein-beta;a major component of seminal fluid and a known prostate cancer biomarker. One of the associated variants is found in the promoter region of the gene, and affects a predicted CREB transcription factor binding site. Gel shift experiments confirm this prediction. We identified the functional basis of the role of variants in the promoter region of the MSMB gene in prostate cancer. We showed that variants in MSMB associated with prostate cancer have a function role in the regulation of this gene. High density genotyping of this locus revealed that the rs10993994 SNP is the most highly associated. The C allele of this variant is part of a functional CREB transcription factor binding site and is associated with high expression of the MSMB gene. Whereas the T allele does not bind CREB and is associated with low expression. This represents one of the first examples of functional characterization of the molecular basis for a genome wide association in cancer. A genome wide association study for breast cancer In collaboration with scientists at the Memorial Sloan Kettering Cancer Center (MSKCC) we completed a whole genome association study of breast cancer in Ashkenazi Jewish subjects with a family history of disease. The scan identified several loci with potentially significant associations. From this data we selected loci for follow up studies in a replication cohort. A locus containing two genes, RNF146 and ECHDC1, was identified as positively associated with breast cancer in Ashkenazi subjects. We subsequently confirmed this association in an independent cohort of Ashkenazi breast cancer subjects. RNF146 has motifs in common with proteins that regulate estrogen receptor. In an assay for estrogen response elements, RNF146 inhibited estrogen response. To further characterize RNF146, we expressed the product in bacterial and mammalian cells. We are collaborating with David Waugh to crystallize the protein to determine its 3-dimensional structure. In addition we are collaborating with Allan Weissman to assess the basis for ubiquitin E3 ligase activity and confirmed the ubiquitin ligase activity. Inhibitors of the Hedgehog/Patched Pathway in Cancer Stem Cells The HH/PTCH pathway has been demonstrated to be mutated in virtually all basal cell carcinomas and a portion of medulloblastomas. In addition, many tumors display ligand-dependent activation of the HH/PTCH pathway including pancreas, prostate, gastrointestinal tract, and small cell lung tumors. To further the development of HH/PTCH inhibitors we designed peptides against TM domains and intracellular loops of the SMO protein. SMO is a member of the G-protein-coupled receptor (GPCR) superfamily, and other GPCRs have been targeted by this approach. While peptides against several TM domains failed to have an effect on cell proliferation, peptides against intracellular loops 2 and 3 were potent inhibitors. The active peptides result in a decrease in expression of HH/PTCH target genes and proteins and antagonize the stimulation of the pathway by purmorphamine, a known pathway agonist. To move these reagents forward into clinical studies we have developed retro inverso D- peptides and performed biodistribution analysis. Radiolabelled peptide was injected into mice and the peptide was found to distribute throughout the body. Efficacy studies are now planned. We have also characterized the methylation of the promoter region of the SMO gene. In cell lines with high expression of SMO, the promoter region is unmethylated;whereas the promoter is methylated in low expressing cells. Retinoblastoma and Pediatric Cancer in Guatemala We have established a collaboration with the principal pediatric onclology hospital in Guatemala the Unidad Nacional de Pediatria (UNOP) in Guatemala City. UNOP is the largest childrens cancer hospital in the country. With colleagues from St. Judes hospital we received permission to submit IRB protocols to the Francisco Maroquin Medical School and assisted them in receiving FWA registration, and received approval for a patient population sample collection and pharmacogenetic study to determine the frequency of variants in the indigenous and admixed populations of such genes as TPMT (involved in azathiopurine metabolism and immunosuppression) and other drug metabolism genes. Sample collection has begun and over 200 samples have been obtained. Epidemiology studies have revealed that the incidence of retinoblastoma is elevated in the capital and that there is an under-ascertainment of rural cases, especially in the several indigenous Mayan populations. This suggests a potential environmental cause for some of the disease. Analysis of acute lymphocytic leukemia (ALL) also reveals significant under-representation of childhood ALL in rural and indigenous regions. Genomics of Renal Cell Carcinoma To further understand the genomic events occurring in renal cell cancers we have undertaken whole exome and whole transcriptome sequencing of renal cancer cell lines and tumors. Using Nimblegen exon capture chips, we have captured all exons and miRNA loci from the 786-0 and SN12C renal cancer cell lines and subjected the resulting DNA to 454 high throughput sequencing. This has resulted in high coverage of most of the exons in the genome and in the identification of variants in a large number of genes. On order to validate these NextGen sequencing results we have taken RNA from the same cell lines and performed whole transcriptome sequencing on the same technology platform. A number of cancer-related genes display high confidence variants in both the genome and cDNA and are candidates for followup analysis on larger sets of tumor DNAs. A number of potential hybrid gene transcripts were also identified, that will be validated in followup analysis. We have also performed Solexa high throughput sequencing on a cDNA library from two primary renal cell carcinoma tumors. This analysis has identified an independent set of variants in cancer-related genes that are being validated in larger numbers of tumor samples. Alternative Splicing and T-cell activation In collaboration with Dr. Richard Lempicki, and Dr. Dawei Huang we have successfully established RT-PCR conditions for analyzing the expression of alternatively spliced transcripts. We have experimentally validated the exon array data by RT-PCR and demonstrated that human CD45, ABLIM1, PRDM1, C1orf21, and BRCA1 genes [summary truncated at 7800 characters]
